Prevalence and risk factors for parental-reported oral health of Inuit preschoolers : Nunavut Inuit Child Health Survey , 2007-2008

Introduction: Studies from the early 20th Century suggest that Inuit had a low prevalence of dental caries. However, Inuit children now experience a high prevalence of tooth decay and dental caries. The main objectives of this study were to provide an estimate of the prevalence and correlates of parental-reported oral health among Inuit preschool-aged children in Nunavut. Methods: Inuit preschool-aged children aged 3 to 5 years from 16 of Nunavut’s 25 communities were randomly selected to participate in the Nunavut Inuit Child Health Survey conducted in 2007 and 2008. The parent/primary caregiver was asked to give written informed consent for their child’s participation. Caregivers were asked to rate their child’s oral and dental health and if their child had any ‘decayed, extracted or filled baby teeth’: an affirmative response designated a child as having reported-caries experience (RCE). Interviewer administered questionnaires included household characteristics, nutritional supplements, past-month qualitative food frequency questionnaire (FFQ), and a 24 hour dietary recall with repeat 24 hour recalls on a 20% sub-sample. Results: The overall participation rate was 72.3% (388 children). Among the participating children, 53% percent were female and the mean age was 4.4 ± 0.9 years. The weighted prevalence of RCE was 69.1% (95% CI: 63.7–74.4%). Caregivers rated their


Introduction
In children under 6 years of age, early childhood caries (ECC) is a condition characterized by the presence of one or more decayed, missing, or filled primary teeth 1 .Among young Canadian Inuit children, the prevalence of ECC has been reported to range from 50 to 97%, but recent statistics are not available [2][3][4][5] .Overall, caries are a considerable problem among Canadian Aboriginal children in contrast to the general population who experience much lower rates of ECC [6][7][8][9][10] .
Early childhood caries has a complex etiology.It is affected by the amount of cariogenic bacteria, the presence of cariogenic foods, and the susceptibility of host tooth 11,12 .
Other environmental and behavioural factors such as oral hygiene habits, socioeconomic status, enamel hypoplasia, fluoride exposure, previous caries experience, access to dental care and other dietary factors also impact caries development [10][11][12] .Early childhood caries may lead to pain, improper speech development, reduced ability to chew, low weight-for-age, lower self-esteem and sleep disturbances; however, the evidence for many of these associations is only weak or moderate 6,13 .
Historical evidence suggests that dental caries were rare among Alaska Natives in the early 20th Century [14][15][16][17] .An Arctic nutrition transition, particularly the reduced intake of traditional food and an increased consumption of high-sugar foods, likely contributed to widespread ECC in Inuit communities 18 .Various studies as early as the 1970s have reported that commercially available market food makes up a significant proportion of the Inuit diet, especially among young children [19][20][21] .In addition, studies among Canadian Inuit adults and Dene/Métis adults and children show that diets higher in market foods are also higher in simple carbohydrates 22,23 .Among Dene/Métis children and Inuit women, percent energy from sucrose was reported to be 18-19% and 12-17%, respectively 22 .It is likely that the Arctic nutrition transition has been exacerbated by other oral health risk factors common in these regions, including the practice of giving bottles to children at bedtime, lack of fluoridated water, inadequate access to dental health care, and lack of education about preventative measures 7,24,25 .
Smaller studies reported that children in Nunavut experience a high prevalence of tooth decay and dental caries.The main objective of this study was to provide the prevalence and correlates of parental-reported oral health of Inuit preschoolers, with a focus on dietary and socioeconomic risk factors.

Study population and ethics
The data presented in the current report were from the Nunavut Inuit Child Health Survey, and details of the methodology have been presented elsewhere 26 .In brief, 16 of the 25 communities in Nunavut were selected to participate in the survey, representing all three regions of the territory.Communities were selected based on region, population size, latitude, and logistic feasibility due to flight routes and financial costs relative to the number of ageappropriate children.Inuit children, ages 3 to 5 years, were randomly selected to participate in the survey.Recruiters were instructed to make three attempts to reach each household with a 3-5 year old.Written informed consent was obtained from the child's primary caregiver, the majority of which were either a biological parent (68.4%) or an adoptive parent (21.6%).Of the 537 households successfully contacted, 75 (11.6%)refused on initial contact and 74 (13.8%) accepted but later cancelled or did not show for the interview, providing a participation rate of 72.3% (388) over the 2 years of data collection.

Questionnaires
After giving written, informed consent, interviews were conducted with parents and primary caregivers about their child's health history, home environment and dietary habits.Because the questionnaires were long, to minimize research burden the oral health history was limited to two questions: 1. How would you rate your child's oral and dental health -very good, good, fair, poor? 2. Does your child have any decayed, extracted or filled baby teeth?
Caregiver's rating of a child's oral health has been used successfully in a previous study with young children 27 .The answers to these questions were given by the parents and caregivers and no objective clinical assessments were made.
When caregivers reported that their child had 'decayed, extracted or filled baby teeth', children were classified as having reported-caries experience (RCE).Finally, caregivers were asked if their child was taking any nutritional supplements such as fluoride, vitamin D or multivitamin and multimineral supplements.Brands were recorded to allow ingredients to be determined.

Dietary intake questionnaires
Each caregiver was asked to complete a qualitative food frequency questionnaire (FFQ) for their child.The FFQ was designed to capture past-month information about common traditional foods that are available in the three regions of

Statistical analyses
Weighted-prevalence rates and 95% confidence intervals (CI) for RCE as well as for caregiver-reported oral health ratings were calculated.From the qualitative FFQ, the percent of children who consumed traditional food, milk and various high-sugar food/beverages was determined.The mean number of times per day these foods were eaten was also calculated.Milk and sugar intake in grams per day was also determined from the 24 hour recall.In univariate and multivariate analyses, the outcome variable of interest was whether or not the child had RCE.Frequencies of milk, high sugar food and traditional Inuit food intake were examined as risk factors using Student's t-test.

Population characteristics
Fifty-three percent (204/388) of the participating children were female and the mean age was 4.4 ± 0.9 years.
Cultural, socioeconomic, and health indicator characteristics have been presented elsewhere 26 .

Food frequency information
Using the qualitative FFQ information, 88.7% (378/380) of caregivers reported that their child drank milk in the past month.Among consumers, the mean frequency of intake was

Twenty-four hour dietary recall information
Milk intake was also higher among children without RCE than with RCE, based on data from the 24 hour recall, which provided quantitative information.Children without RCE drank 325.6 ± 44.8 g/day of milk, whereas children with RCE drank 225.9 ± 17.0 g/day (p = 0.01).However, there were no differences in past-day sugar intake between the two groups (RCE: 109.3 ± 5.3 g/day, no RCE: 110.0 ± 6.6 g/day, p = 0.944).
Socioeconomic characteristics, vitamin/supplement use, and daycare attendance were evaluated (Table 3).Children living in public housing were more likely to have RCE than those living in private housing (74% vs 61.5%, respectively, p <0.05), and children taking vitamins or supplements containing either vitamin D, calcium, or fluoride were less likely to have RCE than children not consuming these vitamins/supplements (60.0%vs 75.5%, respectively, p <0.01).
The multiple logistic regression model predicting an outcome of RCE contained frequency of milk intake, frequency of high sugar food intake, nutrition supplement use, child's age, and housing type (Table 4).In these analyses, taking a fluoride, vitamin D or multivitamin supplement was no longer significant.However, frequency of milk consumption remained significantly associated with RCE (OR = 0.84, 95% CI: 0.73-0.97)where an increase in milk consumption by 1 glass per day reduced the odds of having RCE by 16%.Frequency of high sugar food intake also remained significantly positively associated with the prevalence of RCE (OR = 1.11, 95% CI: 1.02-1.12)where an increase of one item per day of high-sugar food corresponded to an 11% increase in the odds of having RCE.

Discussion
The current study identified a high prevalence of cariesexperience among Inuit preschoolers (69.1%), as reported by parents and caregivers.Studies from the early 1990s in Nunavut report high rates of ECC among young Inuit children 2,3,5 .This is consistent with findings from other Inuit regions.Among Alaska Yup'ik (Natives), 77.4% of 2 to 5 year-olds had caries 28 .Among Inuit in Nunatsiavut Labrador, 97% of the population aged 5 to 22 years had dental caries, and among children 5 to 6 years, 68% were affected by severe tooth decay 4 .Our findings are also consistent with more recent findings from Canadian First Nations communities.Among 3-5 year-old Ontario First Nations' children, 74% had one or more caries 29 .Another randomized control trial conducted in First Nations communities in northern Ontario reported a baseline prevalence of caries among children aged from 6 months to 5 years of 69-73% 25 .
In a study of 408 preschoolers in two disadvantaged and two First Nations communities in Northern and Southern Manitoba, there was a 53.7% prevalence of ECC, as determined by clinical examination 30 .In contrast, the prevalence of ECC among young children in two major Canadian urban areas ranged from 4.6 to 11% 6,7 .Among US pre-schoolers, the prevalence of ECC was 27.9% in 1999-2000 31 .In northwestern Ontario, non-Aboriginal preschoolers experienced a 31.1-35.2%prevalence of ECC, whereas off-reserve Aboriginal preschoolers experienced rates of 73.8-81.6% and rates were even higher among onreserve preschoolers 9 .Overall, caregiver reports have revealed that Inuit preschoolers across Nunavut experience high rates of caries, as is the case in many other Canadian Aboriginal communities, while the general population experience much lower rates.It was noted that when parents classified their child's oral health as 'very good' no RCEs were reported, and that among those who rated their child's oral health as poor, 96.4% had RCEs.In contrast, the terms 'good' and 'fair' were associated with similar prevalence rates of RCEs.As the number of teeth affected by caries was not assessed, it is difficult to determine the validity of the terms 'good' and 'fair' using the prevalence of any RCE.In a small sample of US preschoolers, children whose mothers' gave them an oral-health rating of fair to poor had a significantly higher prevalence of ECC (66%) than those who were given a rating of good to excellent (29%) 32 .In a study evaluating attitudes toward caries in two First Nations communities and two disadvantaged communities in urban areas in Manitoba where the prevalence of ECC is high, 78.3% of caregivers whose children had ECC disagreed that rotten teeth could affect their child's health, whereas only 21.7% of caregivers whose children were caries-free disagreed 30,33 .These findings lead to an interesting hypothesis that a high prevalence of caries in a population may result in caregivers' easier acceptance of sub-standard oral health among preschoolers, which in turn would represent a challenge for dental health and nutritional educational campaigns.
Higher milk intake was independently and significantly associated with having no RCE.In other studies, evidence is available for a protective mechanism of milk for good oral health.In a pediatric population, children with caries, when compared with those without, had lower median intakes of milk at 2-3 years of age 34 .In another study, preschoolers from a Canadian Arctic region had significantly lower odds of severe tooth decay when they drank milk regularly 7 .Milk and milk product consumption was also protective against the development of root caries in an elderly Japanese population 35 .Vitamin D is important in bone mineralization and, although the mechanisms are still unclear, vitamin D was found to affect enamel and dentin mineral densities in a study comparing vitamin D receptor deficient and receptor positive mice 36 .Milk and milk products are also a source of casein and an in-vitro study found that natural casein phosphopeptides contained in yoghurt inhibited demineralization and promoted remineralization of dental enamel 37 .Casein phosphopeptides may be protective by stabilizing calcium phosphate on tooth surfaces.Thus, further research is warranted on the association of milk and/or its constituents with dental health.
While soda pop intake was significantly associated with RCE, sugar-sweetened beverages, chocolate and candy intake were not when considered separately.When all highsugar foods were combined, however, there was a strong independent association between frequency of high-sugar food intake and RCE.It has been reported that children who consume more high -sugar beverages are more at risk for dental caries 7,34 .Sucrose helps stimulate cariogenic bacteria to colonize and adhere permanently to tooth surfaces via α-1-3-rich, water insoluble glucans, allowing acids formed by cariogenic foods easier access to tooth enamel 11 .
Historical literature on Alaska Natives suggests that the prevalence of caries in the early 1900s was low.For example, little evidence of caries was found in teeth from preserved Alaska Native human remains housed at the Smithsonian Institution 14 .Prevalence studies in the 1930s among Alaska Native adults and children also reported low rates of caries [15][16][17] 43 .There is evidence that water fluoridation and fluoride varnishes may reduce the prevalence of ECC among Canadian Aboriginal preschoolers 7,25 .The identification of lower milk intake and higher intake of sugary foods as risk factors for RCE adds to evidence that dietary counselling is also an important component of prevention.Given the success in smoking education messages, where now the majority of homes with preschoolers have restrictions against smoking in the home, this provides some evidence that public health education can have a meaningful impact on health behaviours 26 and that the same success could be achievable for oral health with renewed dental health education efforts developed and shaped by communities.

Limitations
The present study was limited by its cross-sectional design and reliance on parent/caregiver-reported oral health rather than clinically assessed oral health.While the question 'Does your child have any decayed, extracted, or filled primary teeth?' could also potentially include teeth which naturally exfoliated or teeth lost due to trauma, the parental reports cannot be validated with the current data.The strength of the study involves the use of both an FFQ and 24 hour recall to assess dietary habits.However, multiple 24 hour recalls would have strengthened the study by better characterizing usual intake.This was not conducted due to concerns of research burden.

Conclusions
Evidence of a high prevalence of reported caries experience was found among Inuit preschoolers in Nunavut.This helps to confirm the suspected severity of this health problem among young Inuit children.Evidence is reported that low milk and high-sugar food and beverage intake are independently associated with having RCE.This supports the need for continuation of established initiatives in Nunavut that aim to reduce the consumption of soda pop and sugar-sweetened beverages.Inuit-specific information can provide relevant data for communities and public health campaigns aimed at reducing the burden of early childhood caries.
Acceptability for Research Involving Human Subjects was obtained from the McGill Faculty of Medicine Institutional Review Board.A Scientific Research License was obtained from the Nunavut Research Institute.

2 (
Other factors such as nutritional supplement use, daycare attendance and socioeconomic variables, including type of housing (public vs private) and recipient of income support in the household were examined using a χ 2 test; relative risks (RR) and 95% CI were calculated.When risk factors were significantly associated with RCE, they were included in a multiple logistic regression model to examine independent associations.For all analyses, a p-value ≤0.05 was considered significant.Twenty-four hour recall analyses and weighted prevalence estimates were performed in SAS v9.SAS Institute Inc; Cary, NC, USA), while all other analyses were performed using Stata 10 (Stata Corp; TX, USA; http://www.stata.com/stata10/).

Table 1 : Weighted prevalence of reported caries experience and other related oral health indicators among Inuit preschoolers: Nunavut Inuit Child Health Survey, 2007-2008
RCE, Reported caries experience.†As reported by parent or primary care giver.

Table 2 : Differences in mean frequency of milk, high-sugar food and traditional food intake among preschoolers with and without reported caries experience. Nunavut Inuit Child Health Survey, 2007-2008 No. times per day the food or drink was consumed (M ± SE) Food or drink RCE † (n = 254) No RCE (n = 97) P-value ¶
RCE, Reported caries experience.†As reported by parent or primary care giver; ¶ t-tests for differences in means.

Table 4 : Multiple logistic regression to examine risk factors for having reported caries experience, as reported by parent or primary care giver (n = 310)
OR, Odds ratio.
Nunavut has experienced one of the largest increases in dental expenditures in comparison with other regions of Canada, and in 2003 the Government of Canada reported that various oral health programs were in place that involved dietary counselling, oral health promotion, education at the prenatal level as well as in the classroom and well-baby clinics, oral hygiene instruction and the use of chemical sealants 40-42 .Preventative measures are particularly important because the isolation of the communities remains a